Q1) Market model relating rate of return of stock XYZ (denoted RXYZ) to rate of return of a broader index of stock market (denoted RM) is reported as: RXYZ = 4.46 + 0.304RM
|
Coefficient
|
Standard Error
|
|
|
4.460
|
3.0400
|
|
|
0.304
|
0.3243
|
|
|
Analysis of Variance
|
|
|
|
Source
|
DF
|
Sum of Squares
|
|
Regression
|
?
|
141.9
|
|
Residual
|
?
|
3718.9
|
|
Total
|
24
|
?
|
|
|
|
|
Which of given is an suitable null hypothesis?
a) H0: beta1 = 0
b) H0: beta1 = beta0
c) H0: beta1 ≠ 0
d) H0: beta1 ≠ beta0
Q2) What can be done to remedy multicollinearity?
a) Use ANOVA instead of Regression.
b) Drop variables which are correlated from the model.
c) Add dummy variables to prevent problem.
d) None of the above.
Q3) You have been given report which presents 2 regression models with different numbers of independent variables. You don't have the data so you can't recalculate regression
models. However, you have been asked to find out which model is better. How can you
reasonably compare them?
a) Use R-square
b) Use Adjusted R-square
c) Use coefficient of determination
d) You can't compare models with different numbers of independent variables.